DOCSLIB.ORG

Machine Learning Approaches for Outdoor Air Quality Modelling: a Systematic Review

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Machine Learning Approaches for Outdoor Air Quality Modelling: a Systematic Review applied sciences Review Machine Learning Approaches for Outdoor Air Quality Modelling: A Systematic Review Yves Rybarczyk 1,2 and Rasa Zalakeviciute 1,* 1 Intelligent & Interactive Systems Lab (SI2 Lab), Universidad de Las Américas, 170125 Quito, Ecuador; [email protected] 2 Department of Electrical Engineering, CTS/UNINOVA, Nova University of Lisbon, 2829-516 Monte de Caparica, Portugal * Correspondence: [email protected]; Tel.: +351-593-23-981-000 Received: 15 November 2018; Accepted: 8 December 2018; Published: 11 December 2018 Abstract: Current studies show that traditional deterministic models tend to struggle to capture the non-linear relationship between the concentration of air pollutants and their sources of emission and dispersion. To tackle such a limitation, the most promising approach is to use statistical models based on machine learning techniques. Nevertheless, it is puzzling why a certain algorithm is chosen over another for a given task. This systematic review intends to clarify this question by providing the reader with a comprehensive description of the principles underlying these algorithms and how they are applied to enhance prediction accuracy. A rigorous search that conforms to the PRISMA guideline is performed and results in the selection of the 46 most relevant journal papers in the area. Through a factorial analysis method these studies are synthetized and linked to each other. The main findings of this literature review show that: (i) machine learning is mainly applied in Eurasian and North American continents and (ii) estimation problems tend to implement Ensemble Learning and Regressions, whereas forecasting make use of Neural Networks and Support Vector Machines. The next challenges of this approach are to improve the prediction of pollution peaks and contaminants recently put in the spotlights (e.g., nanoparticles). Keywords: atmospheric pollution; predictive models; data mining; multiple correspondence analysis 1. Introduction Worsening air quality is one of the major global causes of premature mortality and is the main environmental risk claiming seven million deaths every year [1]. Nearly all urban areas do not comply with air quality guidelines of the World Health Organization (WHO) [2,3]. The risk populations that suffer from the negative effects of air pollution the most are children, elderly, and people with respiratory and cardiovascular problems. These health complications can be avoided or diminished through raising the awareness of air quality conditions in urban areas, which could allow citizens to limit their daily activities in the cases of elevated pollution episodes, by using models to forecast or estimate air quality in regions lacking monitoring data. Air pollution modelling is based on a comprehensive understanding of interactions between emissions, deposition, atmospheric concentrations and characteristics, meteorology, among others; and is an indispensable tool in regulatory, research, and forensic applications [4]. These models calculate and predict physical processes and the transport within the atmosphere [5]. Therefore, they are widely used in estimating and forecasting the levels of atmospheric pollution and assessing its impact on human and environmental health and economy [6–9]. In addition, air pollution modelling is used in science to help understand the relevant processes between emissions and concentrations, and understand the interaction of air pollutants with each other and with weather [10] Appl. Sci. 2018, 8, 2570; doi:10.3390/app8122570 www.mdpi.com/journal/applsci Appl. Sci. 2018, 8, 2570 2 of 27 and terrain [11,12] conditions. Modelling is not only important in helping to detect the causes of air pollution but also the consequences of past and future mitigation scenarios and the determination of their effectiveness [4]. There are a few main approaches to air pollution modelling—atmospheric chemistry, dispersion (chemically inert species), and machine learning. Different complexity Gaussian models (e.g., AERMOD, PLUME) are widely used by authorities, industries, and environmental protection organizations for impact studies and health risk investigations for emissions dispersion from a single or multiple point sources (also line and area sources, in some applications) [13,14]. These models are based on assumptions of continuous emission, steady-state conditions and conservation of mass. Lagrangian models study a trajectory of an air parcel, the position and properties of which are calculated according to the mean wind data over time (e.g., NAME) [5,14]. On the other hand, Eulerian models use a gridded system that monitors atmospheric properties (e.g., concentration of chemical tracers, temperature and pressure) in specific points of interest over time (e.g., Unified Model). Chemical Transport Models (CTMs) (e.g., air-quality, air-pollution, emission-based, source-based, source, etc.) are prognostic models that process emission, transport, mixing, and chemical transformation of trace gases and aerosols simultaneously with meteorology [15]. Complex and computationally costly CTMs can be of a global (e.g., online: Fim-Chem, AM3, MPAS, CAM-Chem, GEM-MACH, etc.; and offline: GEOS-Chem, MOZART, TM5) and regional (e.g., online: MM5-Chem, WRF-Chem, BRAMS; and offline: CMAQ, CAMx, CHIMERE) scale [16,17]. These models combine atmospheric science and multi-processor computing techniques, highly relying on considerable resources like real-time meteorological data and an updated detailed emission sources inventory [18]. Unfortunately, emission inventory inputs for boundary layers and initial conditions may be lacking in some regions, while geophysical characteristics (terrain and land use) might further complicate the implementation of these models [19]. To deal with complex structure of air flows and turbulence in urban areas Computer Fluid Dynamics (CFD) methods are used [20]. However, recent studies show that the traditional deterministic models struggle to capture the non-linear relationship between the concentration of contaminants and their sources of emission and dispersion [21–24], especially in a model application in regions of complex terrain [25]. To tackle the limitations of traditional models, the most promising approach is to use statistical models based on machine learning (ML) algorithms. Statistical techniques do not consider physical and chemical processes and use historical data to predict air quality. Models are trained on existing measurements and are used to estimate or forecast concentrations of air pollutants according to predictive features (e.g., meteorology, land use, time, planetary boundary layer, elevation, human activity, pollutant covariates, etc.). The simplest statistical approaches include Regression [26], Time Series [27] and Autoregressive Integrated Moving Average (ARIMA) [28] models. These analyses describe the relationship between variables based on possibility and statistical average. Well-specified regressions can provide reasonable results. However, the reactions between air pollutants and influential factors are highly non-linear, leading to a very complex system of air pollutant formation mechanisms. Therefore, more advanced statistical learning (or machine learning) algorithms are usually necessary to account for a proper non-linear modelling of air contamination. For instance, Support Vector Machines [29], Artificial Neural Networks [30], and Ensemble Learning [31] have been applied to overcome non-linear limitations and uncertainties to achieve better prediction accuracy. Although statistical models do not explicitly simulate the environmental processes, they generally exhibit a higher predictive performance than CTMs on fin spatiotemporal scales in the presence of extensive monitoring data [32–34]. Different machine learning approaches have been used in recent years to predict a set of air pollutants using different combinations of predictor parameters. However, with a growing number of studies, it is puzzling why a certain algorithm is chosen over another for a given task. Therefore, in this study we aim to review recent machine learning studies used in atmospheric pollution research. To do so, the remainder of the paper is organized into three sections. First, we explain the method used to select Appl. Sci. 2018, 8, 2570 3 of 27 and scan relevant journal articles on the topic of machine learning and air quality, which conforms to the PRISMA guideline. This section also describes the strategy used to analyze the manuscripts and synthetizeAppl. Sci. the 2018 main, 8, x findings.FOR PEER REVIEW Second, the results are presented and discussed from a general to3 a of detailed 27 and synthetic account. Finally, the last section draws conclusions on the use of machine learning algorithms quality, which conforms to the PRISMA guideline. This section also describes the strategy used to for predictinganalyze the air qualitymanuscripts and theand future synthetize challenges the main of thisfindings. promising Second, approach. the results are presented and discussed from a general to a detailed and synthetic account. Finally, the last section draws 2. Method conclusions on the use of machine learning algorithms for predicting air quality and the future 2.1. Searchchallenges Strategy of this promising approach. Relevant2. Method papers were researched in SCOPUS. The enquiry was limited to this scientific search engine, because it compiles, in a single database, manuscripts which are also indexed in the 2.1. Search
Load more

Recommended publications
  • Part II: Practical Uses of Air Modeling in Litigation And
    Air Modeling as a Tool in Environmental Law and Policy: A Guide for Communities and Environmental Groups Part II: Practical Uses of Air Modeling in Litigation and Regulatory Contexts Clean Air Council 135 South 19th Street Suite 300 Philadelphia, PA 19103 (215) 567-4004 www.cleanair.org September 16, 2016 Authors and Purpose The authors are attorneys at the Clean Air Council at its headquarters in Philadelphia, Pennsylvania. The attorneys working on this White Paper were: Augusta Wilson, Esq., Aaron Jacobs-Smith, Esq., Benjamin Hartung, Esq., Christopher D. Ahlers, Esq., and Joseph Otis Minott, Esq., Executive Director and Chief Counsel. Clean Air Council is a non-profit, 501(c)(3) corporation with headquarters in Philadelphia, Pennsylvania. Acknowledgments The Clean Air Council thanks the Colcom Foundation. This paper would not have been possible without its generous funding and support. The authors would also like to thank Jini Chatterjee, Tessa Roberts and Paul Townsend, students at the University of Pennsylvania Law School’s Environmental Law Project, who conducted invaluable research for this paper, and Sonya Shea, who supervised their work at the Environmental Law Project. Disclaimer This paper is intended as a general introduction to the law and policy of air modeling under the Clean Air Act. Nothing in this paper is intended, nor shall it be construed as creating an attorney-client relationship or providing legal advice. © 2016 Clean Air Council 135 S. 19th Street Suite 300 Philadelphia, Pennsylvania 19103 Table of Contents Introduction .................................................................................................................................1 1. Use of Air Modeling to Support Tort Claims .....................................................................2 a. Challenges to the Use of Particular Modeling Software or to Underlying Assumptions .............................................................................................................2 b.
    [Show full text]
  • Air Dispersion Modeling Guidelines
    Air Quality Modeling Guidelines for Arizona Air Quality Permits PREPARED BY: FACILITIES EMISSIONS CONTROL SECTION AIR QUALITY DIVISION ARIZONA DEPARTMENT OF ENVIRONMENTAL QUALITY November 1, 2019 TABLE OF CONTENTS 1 INTRODUCTION.......................................................................................................... 1 1.1 Overview of Regulatory Modeling ........................................................................... 1 1.2 Purpose of an Air Quality Modeling Analysis .......................................................... 2 1.3 Authority for Modeling ............................................................................................. 3 1.4 Acceptable Models.................................................................................................... 3 1.5 Overview of Modeling Protocols and Checklists ..................................................... 4 1.6 Overview of Modeling Reports ................................................................................ 5 2 LEVELS OF MODELING ANALYSIS SOPHISTICATION .................................. 5 2.1 Screening Models or Screening Techniques ............................................................. 6 2.1.1 Screening Models for Near-Field Assessments ................................................. 6 2.1.2 Screening Techniques for Long-Range Transport Assessments ....................... 7 2.2 Refined Modeling ..................................................................................................... 8 2.2.1 AERMOD .........................................................................................................
    [Show full text]
  • The Role of Natural Variability in Projections of Climate 10.1002/2016GL071565 Change Impacts on U.S
    PUBLICATIONS Geophysical Research Letters RESEARCH LETTER The role of natural variability in projections of climate 10.1002/2016GL071565 change impacts on U.S. ozone pollution Key Points: Fernando Garcia-Menendez1,2 , Erwan Monier2 , and Noelle E. Selin2,3,4 • Natural variability can significantly fl in uence model-based projections of 1Department of Civil, Construction and Environmental Engineering, North Carolina State University, Raleigh, North Carolina, climate change impacts on air quality 2 • Multidecadal simulations or initial USA, Joint Program on the Science and Policy of Global Change, Massachusetts Institute of Technology, Cambridge, 3 condition ensembles are needed to Massachusetts, USA, Institute for Data, Systems and Society, Massachusetts Institute of Technology, Cambridge, identify an anthropogenic-forced Massachusetts, USA, 4Department of Earth, Atmospheric, and Planetary Sciences, Massachusetts Institute of Technology, climate signal in O3 concentrations Cambridge, Massachusetts, USA • It is difficult to attribute the impacts of climate change on O3 to human influence before midcentury or under Abstract Climate change can impact air quality by altering the atmospheric conditions that determine stabilization scenarios pollutant concentrations. Over large regions of the U.S., projected changes in climate are expected to favor formation of ground-level ozone and aggravate associated health effects. However, modeling studies Supporting Information: exploring air quality-climate interactions have often overlooked the role of natural variability, a major source • Supporting Information S1 of uncertainty in projections. Here we use the largest ensemble simulation of climate-induced changes in air Correspondence to: quality generated to date to assess its influence on estimates of climate change impacts on U.S. ozone. F. Garcia-Menendez, We find that natural variability can significantly alter the robustness of projections of the future climate’s [email protected] effect on ozone pollution.
    [Show full text]
  • CMIP5-Derived Single-Forcing, Single-Model, and Single-Scenario Wind-Wave Climate Ensemble: Configuration and Performance Evaluation
    Journal of Marine Science and Engineering Article CMIP5-Derived Single-Forcing, Single-Model, and Single-Scenario Wind-Wave Climate Ensemble: Configuration and Performance Evaluation Alvaro Semedo 1,2,*, Mikhail Dobrynin 3, Gil Lemos 2, Arno Behrens 4, Joanna Staneva 4, Hylke de Vries 5, Andreas Sterl 5, Jean-Raymond Bidlot 6 ID , Pedro M. A. Miranda 2 ID and Jens Murawski 7 1 Department of Water Science and Engineering, IHE-Delft, P.O. Box 3015, 2601 DA Delft, The Netherlands 2 Instituto Dom Luiz, Faculdade de Ciências, University of Lisbon, 1749-016 Lisbon, Portugal; [email protected] (G.L.); [email protected] (P.M.A.M.) 3 Institute of Oceanography, Center for Earth System Research and Sustainability (CEN), Universität Hamburg, 20095 Hamburg, Germany; [email protected] 4 Helmholtz-Zentrum Geesthacht Centre for Materials and Coastal Research, D-21502 Geesthacht, Germany; [email protected] (A.B.); [email protected] (J.S.) 5 Royal Netherlands Meteorological Institute (KNMI), Utrechtseweg 297, 3731GA De Bilt, The Netherlands; [email protected] (H.d.V.); [email protected] (A.S.) 6 European Centre for Medium-Range Weather Forecasts, Shinfield Park, Reading RG2 9AX, UK; [email protected] 7 Danish Meteorological Institute, 2100 Copenhagen Ø, Denmark; [email protected] * Correspondence: [email protected]; Tel.: +31-152-152-387 Received: 22 June 2018; Accepted: 25 July 2018; Published: 1 August 2018 Abstract: A Coupled Model Intercomparison Project Phase 5 (CMIP5)-derived single-forcing, single-model, and single-scenario dynamic wind-wave climate ensemble is presented, and its historic period (1979–2005) performance in representing the present wave climate is evaluated.
    [Show full text]
  • 10.5 Development of a Wrf-Aermod Tool for Use in Regulatory Applications
    10.5 DEVELOPMENT OF A WRF-AERMOD TOOL FOR USE IN REGULATORY APPLICATIONS Toree Myers-Cook*, Jon Mallard, and Qi Mao Tennessee Valley Authority (TVA) ABSTRACT available. For rural sources, the closest representative NWS station may be greater than 50- The Tennessee Valley Authority (TVA) has developed 100 kilometers away. Furthermore, NWS data a WRF-AERMOD tool which transforms the Weather limitations - instrumentation limits, missing data, and Research and Forecasting (WRF) model output into lack of surface parameters required by AERMET - the meteorological input needed by AERMOD have motivated the EPA to explore the use of (AERMAP/AERMET/AERMOD), the EPA- prognostic meteorological models, specifically MM5 recommended model for short-range dispersion (Brode 2008), which can provide a timely and spatially modeling. The tool was used to model 2002 comprehensive meteorological dataset for input to emissions from TVA’s Allen Fossil (ALF) Plant in AERMOD. The EPA has been testing an in-house Memphis, Tennessee. Modeling using NWS data MM5-AERMOD tool for possible use in future was also performed and results were compared. regulatory applications. However, since MM5 is no Analyses of modeling results showed several longer supported by its developers and the WRF differences between the two approaches, with much (Weather Research Forecasting) model is now of the inconsistencies attributed to four main findings. considered the state-of-the-art meteorological model First, the NWS-AERMOD approach tended to stay to replace it, TVA developed a WRF-AERMOD tool more unstable in the summer and stable in the winter that takes WRF three-dimensional meteorological with unrealistically high mechanical boundary layer output (surface and upper air) and transforms it into (MBL) heights.
    [Show full text]
  • Atmospheric Deposition of Pcbs in the Spokane River Watershed
    Atmospheric Deposition of PCBs in the Spokane River Watershed March 2019 Publication No. 19-03-003 Publication Information This report is available on the Department of Ecology’s website at https://fortress.wa.gov/ecy/publications/SummaryPages/1903003.html. Data for this project are available at Ecology’s Environmental Information Management (EIM) website: www.ecology.wa.gov/eim/index.htm, search Study ID BERA0013. The Activity Tracker Code for this study is 16-032. Water Resource Inventory Area (WRIA) and 8-digit Hydrologic Unit Code (HUC) numbers for the study area: WRIAs HUC numbers 56 – Hangman 17010305 57 – Middle Spokane 17010306 Contact Information Publications Coordinator Environmental Assessment Program P.O. Box 47600, Olympia, WA 98504-7600 Phone: (360) 407–6764 Washington State Department of Ecology – www.ecology.wa.gov. Headquarters, Olympia (360) 407-6000 Northwest Regional Office, Bellevue (425) 649-7000 Southwest Regional Office, Olympia (360) 407-6300 Central Regional Office, Union Gap (509) 575-2490 Eastern Regional Office, Spokane (509) 329-3400 Cover photo: Atmospheric deposition samplers on the roof of the Spokane Clean Air Agency Building. Photo by Brandee Era-Miller. Any use of product or firm names in this publication is for descriptive purposes only and does not imply endorsement by the author or the Department of Ecology. Accommodation Requests: To request ADA accommodation, including materials in a format for the visually impaired, call Ecology at 360-407-6764, or visit https://ecology.wa.gov/ accessibility. People with impaired hearing may call Washington Relay Service at 711. People with speech disability may call TTY at 877-833-6341.
    [Show full text]
  • Coupling Multi-Component Models with MPH on Distributed Memory Computer Architectures
    Coupling Multi-Component Models with MPH on Distributed Memory Computer Architectures Yun He and Chris Ding Computational Research Division, Lawrence Berkeley National Laboratory University of California, Berkeley, CA 94720, USA [email protected], [email protected] Abstract A growing trend in developing large and complex applications on today’s Teraflop scale computers is to integrate stand-alone and/or semi-independent program components into a comprehensive simulation package. One example is the Community Climate System Model which consists of atmosphere, ocean, land-surface and sea-ice components. Each component is semi-independent and has been developed at a different institution. We study how this multi-component, multi-executable application can run effectively on distributed memory ar- chitectures. For the first time, we clearly identify five effective execution modes and develop the MPH library to support application development utilizing these modes. MPH performs component-name registration, resource allocation and initial component handshaking in a flex- ible way. Keywords: multi-component, multi-executable, component integration, distributed memory architecture, climate modeling, CCSM, PCM. 1 Introduction With rapid increase in computing power of distributed-memory computers, and clusters of Sym- metric Multi-Processors (SMP) application problems grow rapidly both in scale and complexity. Effectively organizing a large and complex simulation program so that it is maintainable, re-usable, sharable and efficient becomes an important task for high performance computing. Building a comprehensive application system utilizing (and modifying) existing codes developed by different groups is a standard development approach. Component-based software engineering (CBSE) is an emerging trend for software develop- ment in both research and applications.
    [Show full text]
  • Feasibility Study: Modelling Environmental Concentrations of Chemicals from Emission Data
    EEA Technical report No 8/2007 Feasibility study: modelling environmental concentrations of chemicals from emission data ISSN 1725-2237 EEA Technical report No 8/2007 Feasibility study: modelling environmental concentrations of chemicals from emission data Cover design: EEA Layout: Diadeis and EEA Legal notice The contents of this publication do not necessarily reflect the official opinions of the European Commission or other institutions of the European Communities. Neither the European Environment Agency nor any person or company acting on behalf of the Agency is responsible for the use that may be made of the information contained in this report. All rights reserved No part of this publication may be reproduced in any form or by any means electronic or mechanical, including photocopying, recording or by any information storage retrieval system, without the permission in writing from the copyright holder. For translation or reproduction rights please contact EEA (address information below). Information about the European Union is available on the Internet. It can be accessed through the Europa server (www.europa.eu). Luxembourg: Office for Official Publications of the European Communities, 2007 ISBN 978-92-9167-925-6 ISSN 1725-2237 © EEA, Copenhagen, 2007 European Environment Agency Kongens Nytorv 6 1050 Copenhagen K Denmark Tel.: +45 33 36 71 00 Fax: +45 33 36 71 99 Web: eea.europa.eu Enquiries: eea.europa.eu/enquiries Contents Contents Acknowledgements ...................................................................................................
    [Show full text]
  • The State of Climate Modeling in the Great Lakes Basin - a Synthesis in Support of a Workshop Held on June 27, 2019 in Ann Arbor, MI
    The State of Climate Modeling in the Great Lakes Basin A Synthesis in Support of a Workshop held on June 27, 2019 in Ann Arbor, MI Report By: Ontario Climate Consortium Date: September 13, 2019 Acknowledgements: This synthesis report has been prepared in partnership by the Ontario Climate Consortium (OCC), and Environment and Climate Change Canada. The authors wish to acknowledge the following individuals for their contributions to this project: Kristina Dokoska, Ontario Climate Consortium Nishal Shah, Ontario Climate Consortium Greg Mayne, Environment and Climate Change Canada Wendy Leger, Environment and Climate Change Canada Frank Seglenieks, Environment and Climate Change Canada Armin Dehghan, Environment and Climate Change Canada Shaffina Kassam, Environment and Climate Change Canada Heather Arnold, Environment and Climate Change Canada Shivani Vigneswaran, Environment and Climate Change Canada Recommended Citation: Delaney, F. and Milner, G. 2019. The State of Climate Modeling in the Great Lakes Basin - A Synthesis in Support of a Workshop held on June 27, 2019 in Ann Arbor, MI. Toronto, Canada. www.climateconnections.com | 2 Table of Contents Executive Summary .......................................................................................................................................................................... 4 1. Introduction ................................................................................................................................................................................... 7 1.1
    [Show full text]
  • Dispersion V3.23
    Volume 2 Airviro User’s Reference Working with the Dispersion Module How to simulate the dispersion of pollutants Working with the Dispersion Module How to simulate the dispersion of pollutants Amendments Version Date changed Cause of change Signature 3.11 Ago2007 Upgrade GS 3.12 January2009 Upgrade GS 3.13 January2009 Upgrade GS 3.20 May 2010 Upgrade GS 3.21 Dec 2010 Upgrade GS 3.21 June 2012 Review GS 3.22 April 2013 Release GS 3.23 Jan 2014 Upgrade GS 3.23 January 2014 Review GS 3.23 June 2015 Review GS Contents 2.1 Introduction..............................................................................................................7 2.1.1 Why You Need to Use Dispersion Models.........................................................7 2.1.1.1 What’s the Use of Dispersion Simulations.....................................................7 2.1.1.2 How Can Airviro Help?......................................................................................7 2.1.2 Model Assumptions..............................................................................................8 2.1.3 Brief description of the available models..........................................................9 2.1.4 How does Dispersion Module client work?.....................................................18 2.1.5 Guidance for the beginner:................................................................................18 2.1.6 Overview of the Dispersion Module Main Window.........................................19 2.1.6.1 Changing Weather Conditions – Model settings.........................................19
    [Show full text]
  • Atmospheric Rivers in CMIP5 Climate Ensembles Downscaled with a High Resolution Regional Climate Model
    https://doi.org/10.5194/esd-2021-49 Preprint. Discussion started: 22 July 2021 c Author(s) 2021. CC BY 4.0 License. Atmospheric Rivers in CMIP5 climate ensembles downscaled with a high resolution regional climate model Matthias Gröger1, Christian Dieterich1, Cyril Dutheil1, H.E. Markus Meier1,2, Dmitry V. Sein3,4, 1Department of Physical Oceanography and Instrumentation, Leibniz Institute for Baltic Sea Research 5 Warnemünde, Rostock, 18119, Germany 2Research and Development Department, Swedish Meteorological and Hydrological Institute, Norrköping, 601 76, Sweden 3 Shirshov Institute of Oceanology, Russian Academy of Sciences; Moscow, Russia 4 Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research; Bremerhaven, Germany 10 Correspondence to: Matthias Gröger ([email protected]) Abstract Atmospheric rivers (AR) are important drivers of heavy precipitation events in western and central 15 Europe and often associated with intense floods. So far, the ARs response to climate change in Europe has been investigated by global climate models within the CMIP5 framework. However, their spatial resolution between 1 and 3° is too coarse for an adequate assessment of local to regional precipitation patterns. Using a regional climate model with 0.22° resolution we downscale an ensemble of 24 global climate simulations following the greenhouse gas scenarios RCP2.6, RCP4.5, 20 RCP8.5. The performance of the model was tested against ER-I reanalysis data. The downscaled simulation notably better represents small-scale spatial characteristics which is most obvious over the terrain of the Iberian Peninsula where the AR induced precipitation pattern clearly reflect eat-west striking 25 topographical elements resulting in zonal bands of high and low AR impact.
    [Show full text]
  • Evaluation of ISC3, AERMOD, and ADMS Dispersion Models with Observations from Five Field Sites
    EVALUATION OF THE ADMS, AERMOD, AND ISC3 DISPERSION MODELS WITH THE OPTEX, DUKE FOREST, KINCAID, INDIANAPOLIS, AND LOVETT FIELD DATA SETS Steven R. Hanna*, Bruce A. Egan**, John Purdum***, and Jen Wagler**** *CSI MS 5C3, George Mason Univ., Fairfax, VA 22030-4444 USA **Egan Environmental, 75 Lothrop St., Beverly, MA 01915 USA ***17 Peters Way, North Attleboro, MA 02760 USA ****URS Greiner Woodward-Clyde, 38 Chauncy St., Boston, MA 02111 USA ABSTRACT The model evaluation exercise addresses the question whether the new models, ADMS and AERMOD, produce improvements over ISC3 when compared with a range of field observations. ADMS and AERMOD have similar state-of-the-art scientific components, whereas ISC3 contains 1960s technology. The five sets of field observations used in the statistical evaluation represent a cross-section of typical scenarios encountered by modelers. The OPTEX data base concerns non-buoyant tracer releases within an oil refinery complex, and the Duke Forest data base involves non-buoyant tracer releases from area and volume sources in an open field. The Kincaid, Indianapolis, and Lovett data bases all deal with buoyant plumes from tall stacks at power plants. However, the settings are quite different, since the Kincaid plant is surrounded by flat farmland, the Indianapolis plant is located in an urban environment, and the Lovett plant is sited in a valley surrounded by complex terrain with monitors at elevations higher than the stack. Analysis of the model performance measures suggest that ISC3 typically overpredicts, has a scatter of about a factor of three, and has about 33% of its predictions within a factor of two of observations.
    [Show full text]